Impact of Escalation and Cumulative Risks in Chemical Parks

Sunny Choi

5th CCPS China Conference on Process Safety, Nanjing, China

Abstract

Quantitative Risk Analysis (QRA), together with Occupied Building Risk Analysis and Fire and Explosion Risk Analysis, are techniques commonly used by international regulators and facility owners to quantify the risk from process plants. In China, the legislative requirements for risk assessment are stated in Temporary Regulation for Monitoring and Managing of Major Hazardous Installations. The procedure of risk assessment in China is similar to the QRA approaches adopted in other countries such as Singapore, the Netherlands, and the UK where QRA regulations have been in place for several decades as a core safety requirement for hazardous industries.

The traditional approach of QRA focuses on protection against the hazards from a single facility point of view. The potential limitations with such an approach are discussed with hypothetical examples of industrial / chemical parks, similar to those found across China. A cumulative risk assessment approach is proposed to better assess the overall risks from clusters of hazardous facilities. The proposed approach also allows for a cumulative assessment of escalation risk across different plants. Additionally it provides useful information for system level risk mitigation, future planning development and emergency response planning.

As far as the individual facility owner is concerned, a cumulative assessment could also be beneficial for visualizing the actual risk level at a plant – which is not only exposed to hazards from its own facilities but also to hazards from surrounding plants – and thereby ensure adequate protection for employees and assets. The paper also presents an example of how a cumulative analysis approach might be required to determine building protection. In this example, a blast resistant design was found to be necessary even for a non-hazardous installation if it is surrounded by other hazardous plants nearby.